CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Bugs (http://www.cfd-online.com/Forums/openfoam-bugs/)
-   -   Typo in User Guide - Please Check (http://www.cfd-online.com/Forums/openfoam-bugs/65819-typo-user-guide-please-check.html)

jaswi June 26, 2009 18:42

Typo in User Guide - Please Check
 
Dear Developers

Sorry for posting this message here as there is no separate section for reporting such typos.

In the 2nd chapter on page U - 40 under the section 2.1.8.1 following is written:

Quote:

Open the field files for http://www.opencfd.co.uk/openfoam/doc/gif/user103x.png and http://www.opencfd.co.uk/openfoam/doc/gif/user104x.png (0/k and 0/epsilon) and examine the boundary conditions. For a wall boundary condition, http://www.opencfd.co.uk/openfoam/doc/gif/user105x.png is assigned a zeroGradient boundary condition and a fixedValue 0 boundary condition is assigned to http://www.opencfd.co.uk/openfoam/doc/gif/user106x.png. How are the initial values of http://www.opencfd.co.uk/openfoam/doc/gif/user107x.png and http://www.opencfd.co.uk/openfoam/doc/gif/user108x.png arrived at? Positive, non-zero initial values for http://www.opencfd.co.uk/openfoam/doc/gif/user109x.png and http://www.opencfd.co.uk/openfoam/doc/gif/user110x.png must be given to avoid division by 0 in the solution algorithm.
As written here k is to be assigned fixedValue on the walls but this is not correct. The tutorial uses zeroGradient and also the wall functions used in High-Re turbulene models have in-built check to make sure that the boundary conditions on walls are zeroGradient.

Best Regards
jaswi

Ahmed June 29, 2009 14:33

and a fixedValue 0 boundary condition is assigned to http://www.opencfd.co.uk/openfoam/doc/gif/user106x.png
Quote:

Originally Posted by jaswi (Post 220661)
Dear Developers

Sorry for posting this message here as there is no separate section for reporting such typos.

In the 2nd chapter on page U - 40 under the section 2.1.8.1 following is written:

As written here k is to be assigned fixedValue on the walls but this is not correct. The tutorial uses zeroGradient and also the wall functions used in High-Re turbulene models have in-built check to make sure that the boundary conditions on walls are zeroGradient.

Best Regards
jaswi


( and a fixedValue 0 boundary condition is assigned to http://www.opencfd.co.uk/openfoam/doc/gif/user106x.png,) no slip boundary condition, velocity is zero at the wall hence k is 0

jaswi June 29, 2009 17:03

Hi Ahmed

I did not understand your post. By k is 0, do you mean that k being set to fixedValue is correct ?

Please comment on this as after reading your post I tried to set k as fixedValue and value 0 but I get error that at wall k should be zerogradient. The message comes from this section of wallFunctionsI.H:


forAll(patches, patchi)
{
const fvPatch& curPatch = patches[patchi];

if (isType<wallFvPatch>(curPatch))
{
# include "checkPatchFieldTypes.H"
........
.........

where in checkPatchFieldTypes.H it is checked for the patch to be
zeroGradient else raise FatalError. This is what it reads:


if (!isType<zeroGradientFvPatchScalarField>(k_.bounda ryField()[patchi]))
00002 {
00003 FatalErrorIn("wall-function evaluation")
00004 << k_.boundaryField()[patchi].type()
00005 << " is the wrong k patchField type for wall-functions on patch "
00006 << curPatch.name() << nl
00007 << " should be zeroGradient"
00008 << exit(FatalError);
00009 }
00010
00011 if (!isType<zeroGradientFvPatchScalarField>(epsilon_. boundaryField()[patchi]))
00012 {
00013 FatalErrorIn("wall-function evaluation")
00014 << epsilon_.boundaryField()[patchi].type()
00015 << " is the wrong epsilon patchField type for wall-functions on patch "
00016 << curPatch.name() << nl
00017 << " should be zeroGradient"
00018 << exit(FatalError);
00019 }



Any comments are appreciated

BR
jaswi


All times are GMT -4. The time now is 05:29.